The Organometallic Chemistry of N-heterocyclic Carbenes

By Han Vinh Huynh

The Organometallic Chemistry of N-heterocyclic Carbenes describes various aspects of N-heterocyclic Carbenes (NHCs) and their transition metal complexes at an entry level suitable for advanced undergraduate students and above.

The book starts with a historical overview on the quest for carbenes and their complexes. Subsequently, unique properties, reactivities and nomenclature of the four classical NHCs derived from imidazoline, imidazole, benzimidazole and 1,2,4-triazole are elaborated. General and historically relevant synthetic aspects for NHCs, their precursors and complexes are then explained. The book continues with coverage on the preparation and characteristics of selected NHC complexes containing the most common metals in this area, i.e. Ni, Pd, Pt, Ag, Cu, Au, Ru, Rh and Ir. The book concludes with an overview and outlook on the development of various non-classical NHCs beyond the four classical types.

Topics covered include:

• Stabilization, dimerization and decomposition of NHCs
• Stereoelectronic properties of NHCs and their evaluation
• Diversity of NHCs
• Isomers of NHC complexes and their identification
• NMR spectroscopic signatures of NHC complexes
• normal, abnormal and mesoionic NHCs

The Organometallic Chemistry of N-heterocyclic Carbenes is an essential resource for all students and researchers interested in this increasingly important and popular field of research.

• Language: English
• Publisher: Wiley
• Release date: Feb 3, 2017
• ISBN: 9781118698792

Book preview

1

General Introduction

1.1 Definition of Carbenes

According to the International Union of Pure and Applied Chemistry (IUPAC) a carbene [1] is "the electrically neutral species H2C: and its derivatives, in which the carbon is covalently bonded to two univalent groups of any kind or a divalent group and bears two nonbonding electrons, which may be spin‐paired (singlet state) or spin‐non‐paired (triplet state)." In general terms, carbenes are therefore neutral compounds R2C: derived from the parent methylene (H2C:) that feature a divalent carbon atom with only six valence electrons, which result from four bonding electrons in the two R–C bonds and two nonbonding electrons remaining at the carbene center. The geometry at the carbene carbon can be either linear or bent, depending on the degree of hybridization [2]. The linear geometry is based on an sp‐hybridized carbene center with two nonbonding, energetically degenerate p orbitals (px and py). On the other hand, the bent geometry is adopted when the carbene carbon atom is sp²‐hybridized. On transition from the sp‐ to sp²‐hybridization, the energy of one p orbital, usually called pπ, remains almost unchanged, while the newly formed sp²‐hybrid orbital, normally called σ, is energetically stabilized as it acquires partial s character (Figure 1.1). However, the linear geometry is rarely observed, and most carbenes contain a sp²‐hybridized carbene center and are therefore bent.

Schematic illustrating relationship between the carbene bond angle, nature of the frontier orbitals, and singlet–triplet separation, featuring electrophilics.

Figure 1.1 Relationship between the carbene bond angle, the nature of the frontier orbitals and singlet–triplet separation.

For the simple linear case and without considering π contributions of the R‐substituents, only the px¹py¹ electronic configuration is feasible according to Hund’s first rule [3], due to the degeneracy of the px and py orbitals. The two unpaired electrons are both spin up (ms = ½) giving rise to a total spin of S = 1, which in turn results in a spin multiplicity of M = 3 (Equation 1.1). Therefore, linear carbenes are generally in a triplet state. On the other hand, two common electronic configurations are possible for the carbene carbon in bent species. The two nonbonding electrons can singly occupy the two different σ and pπ orbitals with parallel spins (σ¹pπ¹), which also leads to a triplet ground state (³B1). Alternatively, the two nonbonding electrons can also be spin‐paired in the energetically more favorable σ orbital (σ²pπ⁰) leading to a singlet ground state (¹A1).

In addition to the two ground states in bent carbenes, two less favorable electronic configurations are conceivable (not depicted) that give rise to singlet states. The first has two spin‐paired electrons in the pπ orbital (σ⁰pπ², ¹A1), and the second has two electrons singly occupying the σ and pπ orbitals, but with opposite spins (σ¹pπ¹), giving rise to an excited singlet state (¹B1) [4]. The latter two electronic configurations and their states have little significance for the discussion in this work.

The properties and reactivities of bent carbenes are primarily determined by their ground state spin multiplicity [5]. The two singly occupied orbitals in triplet carbenes are unsaturated (open‐shell) and can accommodate one more electrons of opposite spin each. Thus, it is intuitive to assign an electrophilic or diradical character to the carbene carbon (Figure 1.1). Singlet carbenes, on the other hand, contain a fully occupied σ orbital (closed‐shell, nucleophilic) and an empty pπ orbital (electrophilic). The presence of both electrophilic and nucleophilic sites makes singlet carbenes formally ambiphilic.

Equation 1.1

Definition of spin multiplicity for the determination of singlet and triplet state.

Whether a bent carbene adopts the singlet or triplet ground state is determined by the relative energies of the σ and pπ orbitals, which in turn is influenced by the direct substituents R at the carbene carbon. A large energy gap of at least 2 eV (~193 kJ/mol) between the σ orbital and the pπ orbital is required to stabilize a singlet ground state, whereas an energy difference of less than 1.5 eV (~145 kJ/mol) leads to a triplet ground state [6].

The relative energies of σ and pπ orbitals can also be influenced by the steric and electronic effects of the substituents on the carbene carbon atom. For instance, electron‐withdrawing substituents (–I effect) inductively stabilize the σ orbital by enriching its s character and leave the pπ orbital essentially unchanged, thereby increasing the energy gap between the σ and pπ orbitals. Thus the singlet state is favored. On the other hand, electron donating groups (+I effect) decrease the energy gap between σ and pπ orbitals, which stabilizes the triplet state.

Besides inductive effects, which govern the ground‐state spin multiplicity in carbenes, mesomeric effects of the R‐substituents also play a crucial role by influencing the degree of bending in singlet carbenes. If the carbene carbon is attached to at least one π‐accepting group Z (–M effect), for example, Z = COR, CN, CF3, BR2, or SiR3, a linear or quasi‐linear geometry is predicted. In this case, the initial degeneracy of the px and py is broken through π interactions with the Z substituents, therefore allowing for an unusual linear singlet state. On the other hand, π‐electron donating X substituents (+M effect), for example, X = N, O, P, S, and halogens, adjacent to the carbene center increase the energy of the pπ orbital of the carbene carbon atom. Since the σ orbital remains unchanged, the σ–pπ gap is increased, and hence a bent singlet state is favored.

N‐heterocyclic carbenes (NHCs) are carbenes incorporated into heterocyclic rings that must contain at least one nitrogen atom. This minimum structural requirement for an NHC is highlighted in Figure 1.2. In addition to nitrogen, other heteroatoms may also be part of the ring system. Generally, they are bent singlet carbenes. Since they are a centerpiece of this book, the properties and electronic structures of NHCs will be discussed in more detail in Chapter 2.

Schematic illustrating minimum structural requirement for an N-heterocyclic carbene.

Figure 1.2 Minimum structural requirement for an N‐heterocyclic carbene.

1.2 Historical Overview of Carbenes, N‐Heterocyclic Carbenes, and Their Complexes

The following section gives a brief overview on the milestones in the historical development of N‐heterocyclic carbenes and their early use as ligands in organometallic chemistry. It is not the intention to provide a detailed account of short‐lived carbene and carbenoid chemistry in general, which goes beyond the scope of this textbook, the focus of which is classical N‐heterocyclic carbenes. The pioneering work of Jack Hine, William Doering, Philip Skell, Gerhard Closs, and later Robert A. Moss, Wolfgang Kirmse, Hideo Tomioka, and others on such highly reactive species are recounted elsewhere in detail [7], and will not be further mentioned here. Further, this section is divided into two parts. The first deals with the quest for free carbenes, while the second provides a brief summary dealing with the stabilization of carbenes, particularly of NHCs, by transition metals in an historical context. The milestones of these two separate quests are summarized in two timelines depicted in Figure 1.3 and Figure 1.4.

Timeline depicting milestones from 1830 to 2010 in the quest for free carbenes, from Dumas’ attempted methylene synthesis by dehydration of methanol to Hahn’s stable benzimidazolin-2-ylidene.

Figure 1.3 Milestones in the quest for free carbenes.

Timeline depicting milestones from 1900 to 1990 in the quest for carbene complexes, from Tschugaeff’s unrecognized first synthesis of carbene complexes to Schrock’s first metal-alkylidenes.

Figure 1.4 Milestones in the quest for carbene complexes.

1.2.1 The Quest for Free Stable Carbenes

The quest for stable carbenes [8, 9], and in particular that for methylene (H2C:) as the simplest possible representative began back in 1835 [10], when the French chemist Jean‐Baptiste Dumas (1800–1884) tried to synthesize methylene by dehydration of methanol (CH3OH) using phosphorus pentoxide or sulfuric acid. In those days, it was known that carbon can form compounds with oxygen and hydrogen in different oxidation states. In particular, these include carbon monoxide (CIIO), carbon dioxide (CIVO2) and methane (C–IVH4). Thus, the search for a carbon‐hydrogen compound in which carbon adopts the missing intermediate –II oxidation state seemed a reasonable and doable task. In analogy to the aforementioned gaseous carbon compounds, Dumas’ reaction indeed afforded a gaseous compound. Nowadays, we know that he had prepared dimethyl ether (CH3OCH3), which indeed had formed by dehydration, but under condensation of two methanol molecules (Scheme 1.1) instead of from a single molecule.

Schematic structures illustrating Dumas’ attempt to prepare methylene afforded dimethyl ether under condensation of two methanol molecules.

Scheme 1.1 Dumas’ attempt to prepare methylene afforded dimethyl ether.

Pursuing a similar strategy, Anton Geuther (1833–1889) treated chloroform with potassium ethoxide solution in ethanol. He considered chloroform (CHCl3 = Cl2C · HCl) as a simple HCl adduct of dichlorocarbene (Cl2C:), and in 1862, he proposed the formation of the latter according to Scheme 1.2 [11]. This dehydrohalogenation reaction indeed yielded the dichlorocarbene, which under the given reaction conditions was, however, too reactive to be isolated and quickly decomposed to other products.

Schematic structures illustrating dehydrohalogenation of chloroform to give dichlorocarbene, with right arrow labeled KOEt (top) and – KCl, EtOH (bottom) between 2 structural formulas.

Scheme 1.2 Dehydrohalogenation of chloroform to give dichlorocarbene.

In 1895 and 60 years after Dumas report, Johann Ulric Nef (1862–1915) announced that he would tackle the preparation of methylene and its nitrogen free derivatives based on his previous work on derivatives of hydrogen cyanide [12]. He subsequently proposed a theory, in which methylene can be regarded as a suitable building block in organic chemistry and predicted its large scale synthesis [13]. Nevertheless, methylene remained elusive.

The next proposed preparation of a carbene that attracted attention was reported in 1926 by Helmut Scheibler (1882–1966) [14]. According to this report, the reaction of sodium ethoxide with certain esters would—after a complicated and wrong reaction sequence—eventually give tetraethoxyethylene, which further decomposes to two molecules of diethoxycarbene (Scheme 1.3). Interestingly, Scheibler already proposed the formation of a carbene dimer, which dissociates into free monomeric carbenes. This proposal was heavily criticized by the chemical community at that time and finally revoked by later studies.

Schematic structures illustrating wrongly proposed formation of diethoxycarbene, displaying flow from other products to tetraethoxyethylene and then to diethoxycarbene.

Scheme 1.3 Wrongly proposed formation of diethoxycarbene.

Another preparation of the highly reactive dichlorocarbene that attracted the attention of the scientific community was reported in 1960 by Martin Schmeisser (1912–1981), who suggested that treatment of tetrachloromethane with activated carbon at very high temperatures would give dichlorocarbene among a range of other products (Scheme 1.4). The claimed carbene could according to the authors be condensed as a yellow solid in a cold trap immersed in liquefied air. Furthermore, subsequent reactions such as phosgene formation in air and insertions into double bonds were used as indications of the successful dichlorocarbene formation. Further studies by his own team revealed that the activated carbon did not react with CCl4, but catalyzed its decomposition. The yellow product isolated was later correctly identified as a mixture of dichloroacetylene and chlorine.

Schematic structures illustrating proposed dichlorocarbene synthesis by comproportionation of carbon with CCl4.

Scheme 1.4 Proposed dichlorocarbene synthesis by comproportionation of carbon with CCl4.

These numerous failed attempts to isolate free carbenes led the chemical community to think of carbenes only as fleeting intermediates that are too reactive to be isolated [7]. Thus, any further report on their isolation was skeptically regarded. It was during such a time when Hans‐Werner Wanzlick (1917–1988), a former student of H. Scheibler, reported in 1960 that thermolysis of 1,3‐diphenyl‐2‐trichloromethylimidazolidine would lead to an α‐elimination of chloroform and formation of a carbene, which could be isolated as a colorless crystalline material [15]. The molecular weight determined for this material was 300 Da, which is in between the theoretical value for the free carbene and its dimer, and therefore Wanzlick assumed a monomer–dimer equilibrium (Scheme 1.5).

Schematic structures illustrating Wanzlick’s proposed synthesis of an NHC and its monomer–dimer equilibrium.

Scheme 1.5 Wanzlick’s proposed synthesis of an NHC and its monomer–dimer equilibrium.

For Wanzlick’s search for a free carbene, it was unfortunate that he was working on the fully saturated imidazolidine‐based system. Nowadays, we know that the dimeric entetramine form is more favorable for such systems. Although the elimination of chloroform was accepted, Wanzlick’s proposed equilibrium, and therefore the existence of free carbenes in the reaction mixture, was heavily debated.

In 1964, David M. Lemal published results of cross‐coupling experiments that provided evidence against the existence of Wanzlick’s equilibrium and consequently ended Wanzlick’s claim for the isolation of a free carbene [16]. Using NMR spectroscopy, Lemal and coworkers heated a mixture of two tetraaminoethylenes with very similar N‐substituents in xylene under reflux for 2 h, and subsequently subjected the mixture to oxidation using AgNO3 (Scheme 1.6). No oxidized derivatives of cross metathesis products were found, and instead, only symmetrical 2,2′‐bis(imidazolinium) nitrates as oxidation products derived from the two starting materials were identified. Similar cross‐coupling experiments and analysis by gas chromatography by Winberg and coworkers published in 1965 [17], confirmed Lemal’s view that an equilibrium does not exist in the absence of any electrophiles.

Schematic structures illustrating cross-coupling experiments conducted by D. M. Lemal and coworkers by heating a mixture of 2 tetraaminoethylenes and oxidation using AgNO3.

Scheme 1.6 Cross‐coupling experiments conducted by D. M. Lemal and coworkers.

In 1970, Wanzlick again proposed the formation of a free carbene that could be obtained by deprotonation of an imidazolium salt (Scheme 1.7) [18]. Although dimerization of this species was not observed, he unfortunately did not attempt to isolate the compound, which we nowadays know is indeed an N‐heterocyclic carbene. Instead, subsequent reactions with transition metals and other electrophiles were studied with in situ generated carbenes.

Schematic structures illustrating deprotonation of tetraphenylimidazolium salt to give a carbene, with rightward arrow labeled KOtBu, DMSO (top) and – tBuOH, KClO4 (bottom) between 2 skeletal formulas.

Scheme 1.7 Deprotonation of tetraphenylimidazolium salt to give a carbene.

Due to the many failures in isolating stable carbenes, this field of research remained relatively dormant during the next decade, until finally Guy Bertrand and his team in 1988 reported the isolation of the first free carbene that was stabilized by heavier main group elements. Thermolysis or photolysis of (trimethylsilyl)[bis(diisopropylamino)phosphino]diazomethane liberated dinitrogen and a new compound was obtained (Scheme 1.8), which showed properties and behavior similar to phosphaacetylenes, but peculiarly also that of carbenes [19]. The compound was reported to be stable for weeks at room temperature under an inert atmosphere. The ambiguity over whether the isolated compound was a carbene or phaspha‐alkyne was finally addressed by a subsequent paper of the same group, where they provided further and convincing evidence for carbene‐like reactivities of the species in question, such as insertion into double bonds and formation of oxirans with aldehydes [20]. These findings established [bis(diisopropylamino)phosphino]trimethylsilylcarbene as the first stable nucleophilic carbene that has been isolated and fully characterized in substance.

Schematic structures of Bertrand’s [bis(diisopropylamino)phosphino]trimethylsilylcarbene and its relationship (depicted by double-headed arrows) to the respective phospha-alkyne and ylide.

Scheme 1.8 Bertrand’s [bis(diisopropylamino)phosphino]trimethylsilylcarbene and its relationship to the respective phospha‐alkyne and ylide.

Bertrand’s work demonstrated that carbenes can be isolated in pure form. Nevertheless, handling of such carbenes is difficult and requires very elaborate synthetic skills. So, it appears that the chemical community was just waiting for the next discovery, which would propel carbene chemistry from a niche area to mainstream research.

Around the same time as Bertrand’s report on his first stable carbene, industrial chemists at DuPont had identified imidazolin‐2‐thiones as suitable organocatalysts for crosslinking of acid anhydride‐ and epoxide‐functionalized low molecular weight polymers to higher molecular weight polymers for the production of watersoluble coatings for automotive use (Scheme 1.9).

Schematic structures illustrating crosslinking of anhydride- and epoxide-functionalized polymers, with 3 skeletal formulas (arranged horizontally) with rightward arrow labeled catalyst on the bottom.

Scheme 1.9 Crosslinking of anhydride‐ and epoxide‐functionalized low molecular weight polymers.

Since imidazolin‐2‐thiones were not commercially available and older methodologies were too costly for preparation on an industrial scale, a newer cost efficient approach had to be developed. For this purpose, the old Wanzlick‐type chemistry was re‐investigated. The multi‐component condensation reaction of one equivalent of gyoxal, two equivalents of primary amine and one equivalent of formaldehyde provided convenient access to imidazolium chloride salts after addition of hydrochloric acid. Deprotonation of such an imidazolium salt afforded an N‐heterocyclic carbene in situ, which was trapped by addition of elemental sulfur to give the desired imidazolin‐2‐thiones (Scheme 1.10).

Schematic structures illustrating preparation of imidazoline-2-thiones as crosslinking organocatalysts from a skeletal formula involving O, NH2, and R to that involving R, N, and S.

Scheme 1.10 Preparation of imidazoline‐2‐thiones as crosslinking organocatalysts.

The reaction sequence was tested in the laboratory under an inert atmosphere to prevent decomposition of the in situ generated carbene prior to oxidation with sulfur. On an industrial scale using a 2000 L reactor, however, precautions to exclude moisture and air were impossible. Nevertheless, the product yields remained surprisingly high and comparable to those of lab scale experiments. It was therefore concluded that the supposedly reactive carbene intermediate must be relatively stable to survive such conditions.

Subsequently, Anthony J. Arduengo III and his team made the first attempts to isolate free imidazolin‐2‐ylidenes bearing sterically bulky N‐adamantyl wing tip groups, and reported their findings in 1991. The respective imidazolium chloride was treated with sodium hydride (NaH) in tetrahydrofurane (THF) with catalytic amounts of dimethyl sulfoxide (DMSO), which generated the dimsyl anion as an intermediate base (Scheme 1.11). Dihydrogen gas (H2) and sodium chloride (NaCl) formed as easily removable byproducts. Concentration of the THF filtrate afforded large colorless single crystals, which were subjected to single crystal X‐ray diffraction analysis to evaluate the solid state molecular structure of the isolated compound. The results obtained confirmed that deprotonation of 1,3‐di‐1‐adamentylimidazolium chloride had yielded the first free N‐heterocyclic carbene, which is indefinitely stable if kept under an inert atmosphere. Intriguingly, it is also thermally very stable and melts without decomposition at 240 °C [21].

Schematic synthesis illustrating preparation of the stable 1,3-di-1-adamentylimidazolin-2-ylidene (IAd) reacting to THF, DMSO catalyst (left), with structural formula of dimsyl anion (right).

Scheme 1.11 Preparation of the stable 1,3‐di‐1‐adamentylimidazolin‐2‐ylidene (IAd) as the first representative of free NHCs.

Following Arduengo’s successful and seminal isolation of the first free NHC, there have been numerous other reports detailing various methods through which free NHCs can be isolated [22].

In 1995, the groups of Dieter Enders and J. Henrique Teles together reported the isolation and solid state molecular structure of the first 1,2,4‐triazolin‐5‐ylidene. This carbene was prepared by the addition of sodium methoxide to 1,3,4‐triphenyl‐1,2,4‐triazolium perchlorate in methanol, which afforded the neutral 5‐methoxy‐1,3,4‐triphenyl‐4,5‐dihydro‐1H‐1,2,4‐triazole. Upon heating to 80 °C under low pressure, the latter endothermically decomposes under α‐elimination of methanol to form the stable 1,3,4‐triphenyl‐1,2,4‐triazolin‐5‐ylidene (Scheme 1.12) [23]. Apparently, an additional heteroatom in the heterocyclic ring does not negatively affect the stability of the carbenes. Moreover, this carbene should become the first commercially available NHC. Notably, these researchers successfully applied, in principle, the same α‐elimination approach that Wanzlick previously attempted in 1960.

Schematic structure α-elimination of methanol to form the stable 1,3,4-triphenyl‐1,2,4-triazolin-5-ylidene, with 2 rightward arrows labeled NaOMe, MeOH (– NaClO4) and 80 degrees Celsius 0.1 mbar (– MeOH).

Scheme 1.12 Enders α‐elimination approach to the first stable 1,2,4‐triazolin‐5‐ylidene.

In the same year, Arduengo and coworkers also demonstrated that saturated imidazolidin‐2‐ylidenes can be isolated as monomeric species when sufficiently bulky N‐substituents are applied to provide kinetic stability against dimerization to enetetramines.

Thus, by deprotonation of 1,3‐dimesitylimidazolinium chloride with potassium hydride (KH) in dry THF, the researchers obtained the free and monomeric 1,3‐dimesitylimidazolidin‐2‐ylidene (SIMes) as the first example of a saturated NHC (Scheme 1.13) [24]. This carbene was again structurally characterized by single crystal diffraction and showed a melting point of ~ 108 °C.

Schematic synthesis illustrating Arduengo’s isolation of the first imidazolidin-2-ylidene, displaying 2 skeletal formulas and a rightward arrow labeled THF on top.

Scheme 1.13 Arduengo’s isolation of the first imidazolidin‐2‐ylidene.

Finally, in 1999, F. Ekkehardt Hahn and coworkers reported the isolation and solid state structure of the first benzimidazolin‐2‐ylidene, completing the series of free NHCs for the four types of classical NHCs. The authors started with N,N′‐dineopentyl‐substituted ortho‐phenylenediamines, which upon reaction with thiophosgene and triethylamine (NEt3, to trap in situ generated hydrochloric acid) gave N,N′‐dineopentylbenzimidazolin‐2‐thione. Reductive desulfurization using sodium/potassium alloy in toluene yielded the first free benzimidazolin‐2‐ylidene, which was also structurally characterized by single crystal X‐ray diffraction (Scheme 1.14) [25]. Similar to saturated NHCs, the choice of sufficiently bulky wing tip groups is crucial to avoid dimerization to enetetramines.

Schematic structures illustrating Hahn’s approach for the isolation of the first benzimidazolin-2-ylidene, displaying 3 skeletal formulas and 2 rightward arrows alternating them.

Scheme 1.14 Hahn’s approach for the isolation of the first benzimidazolin‐2‐ylidene.

It was now clear that stable NHCs of various backbones can be isolated in pure form. In retrospect, these later discoveries revealed that Wanzlick’s hypothesis on stable NHCs was indeed correct, although he never isolated any by himself. Since the isolation of the first free NHCs 40 years after Wanzlick’s proposal, the chemistry of NHCs has grown exponentially. Nowadays, free NHCs and many of their precursors are commercially available. They have become state‐of‐the‐art organocatalysts and are routine ligands in organometallic chemistry and transition metal mediated catalysis. Applications of NHCs in other areas are beginning to surface, promising an even brighter future for these unique species.

1.2.2 The Quest for Carbene Complexes

Compared to the quest for free carbenes, the historical events related to their stabilization by transition metals, and therefore attempts to isolate carbene complexes, were generally less debated. From today’s point of view, the most obvious route to generate carbene complexes, and generally for all complexes, would be the reaction of a ligand with a chosen transition metal salt. However, such an approach for carbene complexes is historically insignificant, for the first free carbenes were only isolated in 1988 and 1991, respectively. Therefore, most successful pathways fall into the categories of metal‐template‐directed synthesis or in situ generation of free carbene in the presence of metal ions as carbene traps.

The first carbene complex was probably—but also unknowingly—prepared by Lew Alexandrowitsch Tschugajeff [26] (1873–1922) and coworkers in 1915 by the treatment of tetrakis(methylisocyanide)platinum(II) with hydrazine, which afforded a red crystalline complex. Protonation of this red complex with hydrochloric acid led to the release of methylisocyanide and the formation of yellow crystals. For both compounds, he wrongly proposed dimeric PtII species with bridging ligands, which were formally derived from the deprotonation of hydrazine (Scheme 1.15) [27].

Schematic structures illustrating Tschugajeff’s wrongly proposed structures for red complex (left) and yellow complex (right), with a rightward arrow labeled HCl in between.

Scheme 1.15 Tschugajeff’s wrongly proposed structures for his two colored complexes.

Only a reinvestigation of Tschugajeff’s complexes in 1970 by John E. Enemark and coworkers, that also included X‐ray diffraction studies, revealed that these complexes were most likely the first carbene complexes to be synthesized [28]. The red and cationic acyclic monocarbene Pt(II) complex was formed by initial nucleophilic attack of hydrazine on coordinated carbon donors of two isocyanide ligands with concurrent proton shift. Protonation of the iminic nitrogen atom of the red complex with excess hydrochloric acid and isocyanide‐chlorido ligand substitution led to reversible formation of the yellow and neutral dichlorido‐dicarbene PtII complex (Scheme 1.16).

Schematic structures illustrating template-assisted approach to Tschugaeff’s Pt(II) carbene complexes, from [Pt(CNCH3)4]Cl2 to skeletal formula (red complex) and to skeletal formula (yellow complex).

Scheme 1.16 Template‐assisted approach to Tschugaeff’s PtII carbene